The effects of proportional loading, plane stress, and constant thickness assumptions on hydro-mechanical deep drawing process

The goal of this study is to evaluate the effects of proportional loading, plane stress, and constant thickness assumptions on hydro-mechanical deep drawing (HDD) by developing analytical models. The main model includes no simplifying assumption, and then each of the mentioned assumptions is considered in a specific model. The interrelationships between geometrical and mechanical variables are obtained in the finite difference form based on the incremental strain theory, thereby being solved by Broyden algorithm. Published experimental and FE results are used for evaluation of the results obtained in the present work. The results of models under proportional loading, plane stress, and constant thickness conditions show more differences with the experimental data in strain distributions than the results of the main model. Thus, the analyses will be more accurate and the results more reliable without considering such assumptions.     

Optimization of the design parameters of modified die in hydro-mechanical deep drawing using LS-DYNA

The use of a modified die enhances the limiting draw ratio compared to that obtainable in a conventional deep drawing operation. Application of these dies, in conventional deep drawing, eliminated the use of blank holder but enhances the tendency of wrinkling in drawn products. In hydro-mechanical deep drawing process, the punch deforms the blank to its final shape by moving against a controlled pressurized fluid. In this paper, a new concept of the application of modified dies in hydro-mechanical deep drawing is presented. The finite element (FE) simulations of a deep-drawing process using modified dies are performed using the 2-D explicit finite element code LS-DYNA, with the aim of optimization of design parameters and the results are compared with the experimental values. The initial design steps in the design of modified die in finite element simulation were taken from the concept of Tractrix die. The use of Tractrix die enhances the draw ratio but simultaneously increases the tendency of wrinkling. In this paper the design parameters of modified Tractrix die for hydro-mechanical deep drawing are optimized for the successful drawing of cups. It is also experimentally verified that by using such modified dies in hydro-mechanical deep drawing, deeper cups are drawn without wrinkling.     

板液力成形技术及其应用

针对板液力成形原理、特点及其最新技术进行了讨探.从不同的工艺方法入手,介绍了无模液力柔性成形技术、温介质液力成形技术、成对液力成形技术等几种先进的成形工艺,并对其中的关键技术问题进行了分析.该技术在汽车行业的应用可有效地控制变截面板的成形过程,为大幅度降低车身重量、减少能耗,提供了有力的技术支持.     

The effect of normal stress on hydro-mechanical deep drawing process

Normal stress has some role in the deformation analysis of hydroforming processes. In this study, analytical modeling is pursued to evaluate the effect of normal stress on the hydro-mechanical deep drawing (HDD) process. Analyses are carried out for axisymmetric elements of the formed cup-shaped part for increments of the punch travel. The formulations are obtained using mechanical and geometrical relations and the finite difference method, thereby being solved by proper numerical algorithms. Furthermore, in the present work, part thickness is variable, the loading and straining are non-proportional, and bending/unbending effects over the part curvature are considered. The results show that there are some differences between thickness values, radial and circumferential strains and stresses, and punch force under plane stress and three-dimensional stress conditions. Thus, the normal stress should be considered in the design of HDD processes in order to improve accuracy.     

Integration of artificial neural network and simulated annealing algorithm to optimize deep drawing process

Deep drawing is characterized by very complicated deformation affected by the process parameter values including die geometry, blank holder force, material properties, and frictional conditions. The aim of this study is to model and optimize the deep drawing process for stainless steel 304 (SUS304). To achieve the purpose, die radius, punch radius, blank holder force, and frictional conditions are designated as input parameters. Thinning, as one of the major failure modes in deep drawn parts, is considered as the process output parameter. Based on the results of finite element (FE) analysis, an artificial neural network (ANN) has been developed, as a predictor, to relate important process parameters to process output characteristics. The proposed feed forward back propagation ANN is trained and tested with pairs of input/output data obtained from FE analysis. To verify the FE model, the results obtained from the FE model were compared with those of several experimental tests. Afterward, the ANN is integrated into a simulated annealing algorithm to optimize the process parameters. Optimization results indicate that by selecting the proper process parameter settings, uniform wall thickness with minimum thinning can be achieved.     

基于深度强化学习与有限元仿真集成的拉深成形控制

金属板材拉深过程中的压边力是决定成品质量的关键参数,传统压边力控制方法往往需要对高度非线性的拉深过程进行建模,导致其控制结果与实际存在较大偏差。提出一种基于深度强化学习与有限元仿真集成的金属板材拉深过程控制模型,利用深度神经网络强大的预测能力来提取拉深加工过程中的状态信息并进行可靠预测,结合强化学习的决策能力来进行压边力控制策略的学习优化,避免了精确系统动力学模型的拟合以及先验知识的获取。同时,针对板材拉深加工中常见的拉裂质量缺陷与起皱质量缺陷,建立拉深成形性能评价函数,为深度强化学习提供回报信号来指导学习过程,并利用有限元仿真构成深度强化学习的环境模型。试验表明,深度强化学习模型能够有效地进行压边力控制策略优化,有效提高产品质量。所提出的压边力控制模型利用无模型的深度强化学习,能避免拉深过程的系统模型拟合,可提高压边力控制策略的控制效果,同时结合循环神经网络能解决板材拉深加工过程中的部分可观察性问题。     

Inverse technique identification of material parameters using finite element and neural network computation

Experimental identification of anisotropic behavior law is currently obtained by performing several complicated tests and a long duration of experiments. This paper describes a new technique allowing for the identification of HILL anisotropic parameters by inverse technique method based on deep drawing of a cylindrical cup. The identification approach is based on the artificial neural network (ANN) computation trained from finite element simulation. The results obtained by ANN models and by the finite element method shows a good agreement.     

The application of abductive networks and FEM to predict the limiting drawing ratio in sheet metal forming processes

The deep drawing process, one of the sheet metal forming methods, is very useful in the industrial field because of its efficiency. The limiting drawing ratio (LDR) is affected by many material and process parameters, such as the strain-hardening exponent, the plastic strain ratio, friction and lubrication, the blank holder force, the presence of drawbeads, the profile radius of the die and punch, etc. In order to verify the finite element method (FEM) simulation results of the LDR, the experimental data are compared with the results of the current simulation. The influences of the process parameters such as the blank holder force, the profile radius of the die, the clearance between the punch and the die, and the friction coefficient on the LDR are also examined. The abductive network was then applied to synthesize the data sets obtained from the numerical simulation. The predicted results of the LDR from the prediction model are in good agreement with the results obtained from the FEM simulation. By employing the predictive model, it can provide valuable references to the prediction of the LDR under a suitable range of process parameters.     

拉深孔成形的工艺参数优化与实验研究

为了探讨拉深孔成形技术对提高板料成形性能的有效性，以圆筒形件为研究对象，研究了拉深孔成形条件下的成形性能。采用数值模拟、人工神经网络和遗传算法进行板材成形工艺参数优化，得到了最优化的压边力和拉深孔相对密度等拉深工艺参数。根据优化后的结果设计并完成了相关的工艺实验，取得了与有限元模拟相一致的结果，证明了拉深孔成形技术是一种提高板材成形性能的行之有效的方法。     

基于正交试验的半球形拉深成形工艺参数优化

以某半球形TA1钛合金拉深成形件为例,通过ABAQUS有限元软件建立了拉深成形三位模型和数值模拟正交试验方案。以是否破裂和起皱为衡量指标,探究凹凸模间隙、拉深速度和压边力对拉深成形件质量的影响,结果表明:起皱现象主要出现在压边区域不影响成形件质量,对底部最小厚度的影响从大到小依次为拉深速度、压边力以及凹凸模间隙。基于正交试验结果设计优化方案,得到最优工艺参数组合并将其应用于实际生产得到表面质量较好的拉深成形件,结果表明:基于正交试验对拉深成形工艺参数优化的方案可行,可以为企业生产提供指导,具有一定的工程应用价值。     

Prediction of wear behaviour of A356/SiCp MMCs using neural networks

In the present investigation, artificial neural network (ANN) approach was used to predict the wear behaviour of A356/SiC metal matrix composites (MMCs) prepared using rheocasting route. The ANN model was obtained to aid in prediction and optimization of the wear rates of the composites. The effect of the SiC particles size, SiC weight percent, applied pressure and test temperature on the wear resistance was evaluated using the ANN model. The results have shown that ANN is an effective tool in the prediction of the properties of MMCs, and quite useful instead of time-consuming experimental processes.     

Hydrodynamic deep drawing process assisted by radial pressure with inward flowing liquid

A radial pressure can reduce drawing force and increase drawing ratio in hydrodynamic deep drawing. However, conventional hydrodynamic deep drawing cannot attain a radial pressure higher than the pressure in the die cavity. In this research, a modified method, named hydrodynamic deep drawing assisted by radial pressure with inward flowing liquid, was proposed and investigated using both primarily experimental and numerical simulation analysis. A radial pressure higher than the pressure in the die cavity was realized by means of the inward flowing of the liquid during this process. After preliminary experimental validation, FEM was used to explore the forming process. The results from the simulation were compared with those from the experiment. The effects of the radial pressure on the wall thickness distribution, punch force, and compressive stress in the blank flange were studied with assistance of numerical simulation. The process window for radial pressures versus drawing ratios was established in 2Al2O alloy experimentally and cups with drawing ratio of 2.85 were successfully formed.     

Micro-hydromechanical deep drawing of metal cups with hydraulic pressure effects

Micro-metal products have recently enjoyed high demand. In addition, metal microforming has drawn increasing attention due to its net-forming capability, batch manufacturing potential, high product quality, and relatively low equipment cost. Micro-hydromechanical deep drawing (MHDD), a typical microforming method, has been developed to take advantage of hydraulic force. With reduced dimensions, the hydraulic pressure development changes; accordingly, the lubrication condition changes from the macroscale to the microscale. A Voronoi-based finite element model is proposed in this paper to consider the change in lubrication in MHDD according to open and closed lubricant pocket theory. Simulation results agree with experimental results concerning drawing force. Changes in friction significantly affect the drawing process and the drawn cups. Moreover, defined wrinkle indexes have been shown to have a complex relationship with hydraulic pressure. High hydraulic pressure can increase the maximum drawing ratio (drawn cup height), whereas the surface finish represented by the wear is not linearly dependent on the hydraulic pressure due to the wrinkles.     

基于ANSYS/LS-DYNA的板料拉深数值模拟研究

基于连续介质力学及有限元理论,运用动力显式算法建立有限元模型,采用ANSYS/LS-DYNA软件模拟圆形板料的拉深过程,在三种不同加载路径下,研究圆形板料在拉深方向的最大应力和板料厚度的变化规律,以及不同加载方式对板料成形质量的影响,所得结果可为生产实际提供指导.     

镁合金板材液压梯温拉深成形新工艺

结合镁合金板材热拉深工艺的研究状况,根据镁合金常温下塑性差及拉深过程中板材的变形特点,提出新的镁合金板材液压梯温拉深成形工艺,指明新工艺能够提高极限拉深比的本质,并分析了新工艺的可行性。为进一步研究镁合金板料的塑性加工问题及研制合适的液压梯温拉深模具装置奠定技术基础。     

A hybrid NN-FE approach to adjust blank holder gap over punch stroke in deep drawing process

In deep drawing process, the blank holder plays a key role in adjustment of metal flow into the die cavity. Moreover, the quality of drawn parts is extremely affected by this flow. There are two methods of treating the blank holder in deep drawing and its simulation. One is blank holder force (BHF) and the other is blank holder gap (BHG), defined as the fixed distance between the blank holder and the die surface. In previous studies, a large number of experimental techniques have been used to study BHF; however, the amount of theoretical and numerical simulation work to study BHG is insufficient. In the present study, the concept of BHG profile, i.e., variation of BHG over punch stroke is introduced and it is shown that a properly selected BHG profile can improve the section thickness of formed part and result in the drawing of deeper parts. Here, two methods for the optimization of BHG profile are devised, i.e., the local optimization and the global optimization methods. In the first approach, the best BHG in each punch step is determined and finally, the local optimized BHG profile is achieved. In the second method, however, the empirical model for the prediction of final minimum section thickness in terms of BHG profile is obtained using design of experiments and neural networks. In the next stage, the proposed model is implanted into a simulated annealing optimization procedure to identify a proper BHG profile that can produce the desired blank thickness. Afterward, the BHG profile approach is applied to a variety of initial thicknesses, blank diameters, and materials in order to examine the robustness of method. In this paper, ABAQUS finite element package is used to gather finite element (FE) data and several experiments are performed to verify the FE results.     

神经网络技术在薄板拉深成形智能化过程中的应用

应用Matlab中内建的神经网络工具箱建立了三层前向反馈BP神经网络模型,并利用拉深试验中采集数据样本集对模型进行前期学习训练,达到指定目标误差后再利用另外一些实际试验样本集来验证所建模型。结果表明：此三层BP模型模拟计算结果与试验结果的相对误差在1%之内,可有效地预测薄板拉深成形过程中的成形性能参数设置是否合理,从而为实现薄板拉深成形过程的智能化预测奠定一个基础。     

球形件反复拉深成形工艺有限元模拟及试验

提出了球形件反复拉深成形工艺,对该工艺进行了有限元模拟与试验验证。与一次拉深成形工艺进行了对比分析,结果表明：凹模圆角区与法兰区应力状态及大小基本相同,但在球底区,反复拉深时径向应力和周向应力都远小于一次拉深的应力,球底区径向应力和周向应力基本为压应力或很小的拉应力;反复拉深时球底区厚度方向应变明显减小,一次拉深、二次反复拉深、三次反复拉深成形的制品最薄点减薄率分别为0.189、0.122、0.049,三次反复拉深可实现近等壁厚制品的拉深成形。该工艺与筒底冷校形工艺相结合,可实现近等壁厚深筒形零件的拉深成形。     

应用TAGUCHI法对微拉深成形有限元模拟结果的误差分析

为考察基于传统本构模型和求解算法的有限元程序模拟金属微成形过程的适应性,设计正交试验,应用自行开发的板料成形有限元程序ARVIP-3D对微成形和等比放大10倍的常规成形进行模拟计算,并根据质量工程中的TAGUCHI法,参考信噪比的概念,对材料参数和工艺条件等因素对两类相似拉深的模拟结果误差进行了分析。结果表明：微成形中的尺度效应主要是成形机理与常规成形不同而引起的,基于传统本构模型和求解算法的有限元程序不能正确模拟微成形过程,但可以模拟出摩擦引起的尺度效应;宏微观尺度的成形模拟结果误差主要来自于浮点运算的误差,但几何和力学条件的处理也给模拟误差带来了影响。因此,在开发适合于微成形模拟的新型有限元程序中,不仅要考虑本构模型的特殊性,几何和力学条件的相关处理算法也应进行改进。     

盒形件无模充液拉深液压加载路径的研究

凹模型腔内的液体预胀形压力和工作压力是决定充液拉深过程中能否拉深出合格零件的重要工艺参数.通过采用Dynaform软件,对无模充液拉深盒形件工艺进行数字仿真,就液体压力参数对板料成形性能影响进行分析与研究.结果表明,合理地控制凹模型腔内液体加载路径和压力参数可以有效地提高板料的成形性能和成形件质量.并由液体压力所带来的摩擦保持效应可使成形件壁厚分布均匀,一次拉深的盒形件最大相对高度可达到4.25.